The Artima Developer Community
Sponsored Link

Weblogs Forum
Love Constraints

25 replies on 2 pages. Most recent reply: Oct 31, 2006 7:56 AM by Achilleas Margaritis

Welcome Guest
  Sign In

Go back to the topic listing  Back to Topic List Click to reply to this topic  Reply to this Topic Click to search messages in this forum  Search Forum Click for a threaded view of the topic  Threaded View   
Previous Topic   Next Topic
Flat View: This topic has 25 replies on 2 pages [ « | 1 2 ]
Brian Ivie

Posts: 1
Nickname: briani
Registered: Oct, 2006

Re: Love Constraints Posted: Oct 24, 2006 4:14 PM
Reply to this message Reply
Advertisement
The biggest concern with the “release early/release often” mantra is that it too often falls victim to the “nickle-and-dimed-to-death” problem:

“We really should do [major architecture change] but we don’t have time before [incremental release date]. So instead we will do [bandaid solution] and ‘fix it later.’” A year or two later: “If we had known that we would have so many problems caused by [cumulative of many bandaid solutions] we would have taken the time to do [major architecture change], but now we have too much invested to make any major changes.

I think customer input is very important, but if it is paramount the results can be disastrous. I would cite Windows ME as an example of focusing too much on what the customer wants, without the proper technical balance. I think balance is the keyword. The solution is not anything “extreme.” Listen early and often to the customer, but balance their interests and concerns with architectural concerns.

To cite yet another book: “Zen and the Art of Motorcycle Maintenance”: Good programming requires “great peace of mind.”

Isaac Gouy

Posts: 527
Nickname: igouy
Registered: Jul, 2003

Re: making movies Posted: Oct 25, 2006 9:01 AM
Reply to this message Reply
Jeff Ratcliff wrote
> I wasn't my intention to criticise the book as a whole,
> but to criticise one idea that you claim was in it. Merely
> stating the Joe Blow has a different opinion in some book
> without elaborating on Joe's point or linking to it is not
> much of an argument and shame on me for giving it more
> credibility than it deserves. I promise I won't make that
> mistake again.

"not much of an argument"
Nor was it intended to be, nor did it need to be in response to your assertion that "Writing software isn't like making a movie". It was simply to say that other people might have a different opinion about that.

Isaac Gouy

Posts: 527
Nickname: igouy
Registered: Jul, 2003

Messiah Posted: Oct 25, 2006 9:09 AM
Reply to this message Reply
> >> However I agree that references to Händel's Messiah or
> > other artistic enterprises don't really further our
> > understanding of software or software projects.
>
> To the contrary, I'd argue that observing the habits of
> any highly productive person can benefit developers and
> software projects. In these specific examples, several
> observations are relevant to software projects:
>
> * Both composers got paid for results, not for the number
> of hours spent or a fixed salary.
>
> * They had to meet non-negotiable deadlines.
>
> * Their work was ultimately judged by their "users"
> (audience), not by their peers. Thus, they had to be
> concerned about the UI the most (the effects the work
> would create on the audience), and not so much about the
> techniques or tools to achieve those effects.
>
> * They had to choose techniques and tools with execution
> in mind: The performers had to be able to easily execute
> what the score instructed them to do, i.e., they had to
> use familiar techniques.
>
> * In these examples, they did not aim to break new ground,
> but rather to meet the objectives of (a) pleasing their
> "users", and (b) meeting their deadlines. Instead of
> re-inventing the wheel, they re-used much earlier
> material, and stayed within the confines of well-known
> frameworks.
>
> * While they followed conventional frameworks, they added
> flourish and originality at certain places to improve
> "usability" (to please their audience).
>
>
> These are just a few observations in this context, and the
> way it applies to developers and architects are rather
> common-place restatements of what has already been said so
> many times and in so many books and blogs:
>
> * If you project's success is judged by how pleased users
> are with it, focus on usability, and don't worry about
> what techniques or frameworks to use to achieve those
> effects.
>
> * If you want to complete your project in time, don't
> break new ground, but simply imitate what has worked in
> the past, step-by-step. Follow a well-oiled formula, i.e.,
> use a framework.
>
> * Add your individual touch or innovation in well-chosen
> spots, and only to improve usability (to please users).
>
> * When you architect something, keep in mind how easy
> implementation is for developers. Choose a solution that's
> easier to implement over one that's more novel.
>
> Again, these notions have been rehashed time and again.


You're forcing trite generalizations from your pre-conceived notions about software development back onto a different context.

Jeff Ratcliff

Posts: 242
Nickname: jr1
Registered: Feb, 2006

Re: making movies Posted: Oct 25, 2006 11:13 AM
Reply to this message Reply
> Nor was it intended to be, nor did it need to be in
> response to your assertion that "Writing software isn't
> like making a movie". It was simply to say that other
> people might have a different opinion about that.

Somebody might have a different opinion than me? Such a thing would never have occurred to me if you hadn't pointed it out.

Now that I understand the possibility that other opinions might exist, I wouldn't want you to waste your time in the future directing me to other opinions unless you had some substantive point to make.

Matt Gerrans

Posts: 1153
Nickname: matt
Registered: Feb, 2002

Re: So much wrong Posted: Oct 26, 2006 10:43 AM
Reply to this message Reply
> Writing software isn't like making a movie - Art doesn't contain bugs.

Funny. Watching X-Men 3 a week or two ago, I was amused by a scene where Wolverine is fighting some guy who shoots several weird spike things into him. Of course, Wolvie pulls them out, quickly heals and flays the guy, but his white shirt has big holes and lots of blood. Then in the next scene (immediately after, still out in some forest) his shirt is nice and clean and undamaged. Is that not a bug in the movie?

Jeff Ratcliff

Posts: 242
Nickname: jr1
Registered: Feb, 2006

Is it reproducible? Posted: Oct 26, 2006 5:36 PM
Reply to this message Reply
> Is that not a bug in the movie?

Can you have the guy impale Wolverine again so we can see if the shirt is still clean?

Seriously, it's a continuity error, not a bug. Correcting such a problem isn't like correcting a bug at all.

If there's a bug in software you have to examine the code or set breakpoints to track it down before you can correct it.

If there's a continuity error in a movie, no research is required. There is no hidden state or algorithm, all the needed information is right before your eyes.

If software were really like a movie, having each feature work correctly at least once would always mean the product would be ready to ship.

Isaac Gouy

Posts: 527
Nickname: igouy
Registered: Jul, 2003

Like, I mean like... Posted: Oct 28, 2006 9:05 AM
Reply to this message Reply
Jeff Ratcliff wrote

> Seriously, it's a continuity error, not a bug. Correcting
> such a problem isn't like correcting a bug at all.
-snip-
> If software were really like a movie ...

Your comments seem to be based on a very narrow definition of the word 'like', the dictionary at hand includes these meanings
- having some or all of the qualities ...
- resembling in some way ...

Jeff Ratcliff

Posts: 242
Nickname: jr1
Registered: Feb, 2006

How can one guy be so insightful! Posted: Oct 28, 2006 8:11 PM
Reply to this message Reply
1> Your comments seem to be based on a very narrow definition
> of the word 'like', the dictionary at hand includes these
> meanings
> - having some or all of the qualities ...
> - resembling in some way ...

Gee, another very helpful post. To think that I've been using the word "like" for almost 50 years and I never would have understood its true meaning if you hadn't informed me.

Cameron Purdy

Posts: 186
Nickname: cpurdy
Registered: Dec, 2004

Re: So much wrong Posted: Oct 29, 2006 2:27 PM
Reply to this message Reply
> Funny. Watching X-Men 3 a week or two ago, I was
> amused by a scene where Wolverine is fighting some guy who
> shoots several weird spike things into him. Of course,
> Wolvie pulls them out, quickly heals and flays the guy,
> but his white shirt has big holes and lots of blood.
> Then in the next scene (immediately after, still out in
> in some forest) his shirt is nice and clean and undamaged.
> Is that not a bug in the movie?

Movies are fault tolerant. ;-)

Peace,

Cameron Purdy
http://www.tangosol.com/

Cameron Purdy

Posts: 186
Nickname: cpurdy
Registered: Dec, 2004

Re: Love Constraints Posted: Oct 29, 2006 2:31 PM
Reply to this message Reply
> If I were given less time to work on a project, I would
> leave. I wonder what universe 37 signals hails from that
> they make such broad unsubstantiated claims and have
> legions of people ready to jump and nod their way to
> orgasm in agreement. This reminds me, I wish Hani
> Suleiman blogged more.

I too wish Hani blogged more, but I actually understood where this blog was coming from. It seems to me that we are often being required to do an impossibly large number of things in an impossibly small time, and just to make it exciting, it seems that we get extra curve balls / googlies thrown at us all throughout.

The forces of impossibility tend to force prioritization and focus that a lack of constraints would not. As but one example, we (at Tangosol) are fond of saying that we were lucky when we tried to get venture capital and failed, because if we had succeeded, we would have never been forced into the areas from which our Coherence product emerged. So constraints are a PITA, but they do act as natural steering mechanisms, and they can be very positive for a company.

My $.02 anyhow ;-)

Peace,

Cameron Purdy
http://www.tangosol.com/

Achilleas Margaritis

Posts: 674
Nickname: achilleas
Registered: Feb, 2005

Re: Love Constraints Posted: Oct 31, 2006 7:56 AM
Reply to this message Reply
Software is not art, it is engineering. And since it is engineering, it must have strict laws. Art is something that does not have real world consequences when used; it is a totally different thing from software.

Flat View: This topic has 25 replies on 2 pages [ « | 1  2 ]
Topic: Love Constraints Previous Topic   Next Topic Topic: What is the type of 42?

Sponsored Links



Google
  Web Artima.com   

Copyright © 1996-2019 Artima, Inc. All Rights Reserved. - Privacy Policy - Terms of Use